Extracting images from PDFs

gistfile1.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import sys

pdf = open(sys.argv[1], "rb").read()
minimum_seek = 20
startfix = 0
endfix = 2
i = 0

formats = {
    "jpeg": {
        "start": b'\xff\xd8',
        "end": b'\xff\xd9'
    }
}

filenumber = 0
while True:
    istream = pdf.find(b'stream', i)
    if istream < 0:
        break
    print(istream)
    iend = pdf.find(b'endstream', istream)
    if iend < 0:
        raise Exception("Didn't find end of stream!")

    istart = pdf.find(formats["jpeg"]["start"], istream, istream + minimum_seek)
    if istart < 0:
        iend = pdf.find(b'endstream', istart)
        data = pdf[istream:iend]
        datafile = open("data%d" % filenumber, "wb")
        datafile.write(data)
        datafile.close()    
        i = istream + minimum_seek
        filenumber += 1
        continue
    iend = pdf.find(formats["jpeg"]["end"], iend - minimum_seek)
    if iend < 0:
        raise Exception("Didn't find end of JPG!")

    istart += startfix
    iend += endfix
    print("JPG %d from %d to %d" % (filenumber, istart, iend))
    jpg = pdf[istart:iend]
    jpgfile = open("jpg%d.jpg" % filenumber, "wb")
    jpgfile.write(jpg)
    jpgfile.close()
     
    filenumber += 1
    i = iend

Credit to http://nedbatchelder.com/blog/200712/extracting_jpgs_from_pdfs.html
I ported it to Python3 and am working on extracting other image types.

I used this a couple years ago in python 2 and it worked. Recently I tried this again in both python 2 and python 3, but after a few loop, it raise Exception("Didn't find end of stream!"). It seems to me there is some change with pdf. Do you think so?